Introduction
In the past few decades, it has been characteristic for gerontological theory to divide old age into ever smaller and more distinguishable life stages (Degnen Reference Degnen2007). Prolonged lifespans, improved health and economically more secured retirement in Western societies have created heterogeneity among older adults, and active retired people poorly fit the negative stereotypes of old age. Instead of one homogeneous ‘old age’, different groups and life stages have been conceptualised through categorisations such as the ‘young old’ and the ‘old old’ (Myerhoff Reference Myerhoff, Kertzer and Keith1984), the ‘oldest old’ (Featherstone and Hepworth Reference Featherstone, Hepworth, Bytheway, Keil, Allat and Bryman1989), ‘advanced old age’ (Heikkinen Reference Heikkinen2004) or the ‘frail elderly’ (Kayser-Jones Reference Kayser-Jones1981). In its clearest form, the internal division within old age has been expressed in the theory of the third age.
The third age is depicted as a phase of new freedom and opportunities after retirement, active lifestyle, and is characterised by good physical and mental condition (Laslett Reference Laslett1989). Opposite to this is the fourth age, which represents ‘real old age’, and has been described in terms of a radical decline in functional ability, disease, frailty, and an increased need for outside help and medical assistance. The fourth age has been referred to as a period of final dependence, decrepitude and death (Laslett Reference Laslett1989). Gilleard and Higgs (Reference Gilleard and Higgs2013) have pointed out that in the social imaginary of age and ageing, the fourth age represents feared characteristics: otherness, abjection, frailty and marginalisation. It seems to form an abject class: the figure of a terminal place from which there is no escape (Gilleard and Higgs Reference Gilleard and Higgs2011a ). What inevitably moves people to the fourth age is the failure of self-management, which is then replaced by institutional forms of care (Gilleard and Higgs Reference Gilleard and Higgs2010). This ‘terminal destination’ is thus a cultural location stripped of the social and cultural capital that is most valued and which allows for the articulation of choice, autonomy, self-expression and pleasure in later life (Gilleard and Higgs Reference Gilleard and Higgs2010). The fourth age is thus seen as a cultural category rather than a chronological term (Minichiello, Browne and Kendig Reference Minichiello, Browne and Kendig2000), described by increasing prevalence of frailty as well as various forms of psychological mortality, i.e. loss of identity, psychological autonomy and a sense of control (Baltes and Smith Reference Baltes and Smith2003). According to Lloyd et al. (Reference Lloyd, Calnan, Cameron, Seymour and Smith2014), previous studies hold that it is loss of agency that makes a fourth ager. Autonomy is thus a major concept to describe the difference between the third and fourth age.
In ageing research, the concept of autonomy is usually associated with independent living (Beswick et al. Reference Beswick, Rees, Dieppe, Ayis, Gooberman-Hill, Horwood and Ebrahim2008; Horgas, Wilms and Baltes Reference Horgas, Wilms and Baltes1998; Portacolone Reference Portacolone2011), health and functional ability (Ozaki et al. Reference Ozaki, Uchiyama, Tagaya, Ohida and Ogihara2007; Perrig-Chiello et al. Reference Perrig-Chiello, Perrig, Uebelbacher and Stähelin2006), mobility (Leveille et al. Reference Leveille, Penninx, Melzer, Izmirlian and Guralnik2000; Queniart and Charpentier Reference Queniart and Charpentier2011; Schwanen and Ziegler Reference Schwanen and Ziegler2011), and economic and material independence (Plath Reference Plath2008; Schafer, Mustillo and Ferraro Reference Schafer, Mustillo and Ferraro2013). In his book Dependence and Autonomy in Old Age, Agich (Reference Agich2003: 6) discusses varying conceptualisations of autonomy and concludes that ‘autonomy is taken to be equivalent to liberty, self-rule, self-determination, freedom of will, dignity, integrity, individuality, independence, responsibility and self-knowledge; it is also identified with the qualities of self-assertion, critical reflection, freedom from obligation, absence of external causation and knowledge of one's own interest’. Wide conceptual consideration of this kind is particularly important in relation to the third and fourth age and the perceptions of autonomy among the oldest people.
Several studies have shown that older adults seek to distance themselves from the category of old, and maintain a more youthful identity of the ‘not old’. These studies have also found that the perceptions of autonomy play a central role in people's own conceptions of age categorisations. In Hurd's (Reference Hurd1999) interview and observation study at a seniors' centre in Canada, 50–90-year-old members' concept of the ‘not old’ did not refer to any distinct chronological category. Instead it was based on a positive, active and youthful image of the self and body, which distanced them from the ‘olds', who were characterised by living in a nursing home, having serious health problems and not being capable of functioning as energetic individuals. Degnen (Reference Degnen2007) found in her ethnographical study in a seniors' centre in Britain that members elaborated oldness by constant monitoring of each other's physical and mental statuses, such as the capability to attend activities independently in the community centre, ability to remember and concentrate, and consistency of narrative. Any perceived decline in other individual members circulated through gossip groups. In an Australian interview study with 65–89-year-old people, Minichiello, Browne and Kendig (Reference Minichiello, Browne and Kendig2000) found that their interviewees made a distinction between those who are ‘still trying’ and those who are ‘no longer trying’ to participate in life, remain healthy, engage in meaningful activity, maintain their physical presentation and remember things. In a study by Townsend, Godfrey and Denby (Reference Townsend, Godfrey and Denby2006), older people classified their age-peers as heroes/heroines if they managed to keep going regardless of ageing, villains if they were giving up and victims if they had lost their life management due to health problems. The latter studies thus evince that in addition to physical and mental functioning, the maintenance of the ‘not old’ status is thought to require a certain active state of mind, i.e. willingness to fight ageing. They simultaneously show that older adults' descriptions of old age involve moral judgements against those who do not fight ageing.
These studies have thus evinced that in various contexts people use rich argumentation to distance themselves from the category of the ‘old’ as it locates an individual in a marginalized and discriminated position within age relations (Calasanti Reference Calasanti2004). Therefore, people flexibly (re)interpret their own capabilities in order to avoid the social category of the old (Lloyd et al. Reference Lloyd, Calnan, Cameron, Seymour and Smith2014). The studies also highlight that the ‘otherness' of old age is not always based on chronological age but on the ability to live an independent and autonomous life. The others are the ‘ability-others' who have lost their autonomy and therefore represent the true old age. Importantly, the studies also show that the third and fourth age do not effectively reflect the categories that people use when identifying themselves and others (Lloyd et al. Reference Lloyd, Calnan, Cameron, Seymour and Smith2014). Rather, there seems to be a certain ‘grey’ zone between these life stages, and this borderline is under consideration when people think and talk about their autonomy.
In this article, our aim is to explore constructions of autonomy among community-dwelling Finnish 90–91-year-old men and women, based on thematic interviews. As a group, they can be seen as both having and not having features that represent the fourth age. In a chronological sense they belong to the group of the oldest, most of them have various functional limitations hampering their daily lives and many of them receive outside assistance with daily tasks. On the other hand, community-dwelling nonagenarians have maintained their autonomy by being able to live at home instead of in an institution, and by preserving their decision-making abilities. Of course, the group of the oldest old includes a substantial degree of diversity, and does not therefore represent a homogeneous group of people. However, people at the age of 90 are aware that, regardless of their physical and mental condition and living environment, they are close to the final frontiers of autonomy. It is thus theoretically important to analyse the ways in which people in this group consider the boundaries of autonomy at their age, and how this relates to the ways in which they identify themselves in terms of age-based categories. In the analysis, we also pay attention to potential gender-based differences in constructions of autonomy.
Material and methods
The data-set is part of the Vitality 90+ study, which is a multi-disciplinary research project which started in 1995. The Vitality 90+ study consists of longitudinal survey studies, life-story interviews, tests of functional ability and laboratory tests with 90 years old or older people living in Tampere, southern Finland. The study connects the medical and social sciences with gerontology and geriatrics (Enroth et al. Reference Enroth, Raitanen, Hervonen and Jylhä2013; Jylhä et al. Reference Jylhä, Paavilainen, Lehtimäki, Goebeler, Karhunen, Hervonen and Hurme2007; Lisko et al. Reference Lisko, Tiainen, Stenholm, Luukkaala, Hervonen and Jylhä2011).
The data-set analysed in this paper comes from life-story interviews with 45 community-dwelling people born in 1921 and 1922 carried out in the Vitality 90+ study in 2012. There were 695 people in that age cohort living in Tampere in 2012. A request for an interview and a short questionnaire were sent to every fifth woman and man. Only 25 per cent of the women were willing to give an interview while the percentage among men was 48 per cent. This might be a result of men at this age being generally in better condition compared to women (Berg et al. Reference Berg, Hassing, McClearn and Johansson2006). The interview participants included 25 women and 20 men aged 90 or 91 years. All the interviewees gave their informed consent before the interview.
Among the interviewees, 29 lived alone, 11 with a spouse and five lived in some kind of senior housing; 24 respondents reported that they did not need any outside help with light housework, 14 needed help sometimes and seven needed daily help. Self-rated health was most often average (27 persons), while 15 considered their health good or fairly good, and three poor.
In thematic interviews, the informants were first asked to tell their life story from childhood to the present day in their own words. Additional questions concerned the interviewees' health, retirement, hobbies and housing, as well as their views on longevity and older people's status in society. The shortest interview lasted 34 minutes, and the longest three hours and 20 minutes. Most interviews took between 90 minutes and two hours. The interview guide did not include any explicit questions concerning interviewees' perceptions of their autonomy. Therefore, this analysis is based on the entire data-set (totalling 1,200 pages).
We used membership categorisation analysis (MCA) as our method. MCA was originally created by Harvey Sacks (Reference Sacks and Jefferson1992a , Reference Sacks and Jefferson1992b ) based on Harold Garfinkel's (Reference Garfinkel1967) ethnomethodology. The basic idea of MCA is to study how people include and exclude themselves and others from different social categories, and how they attach various qualities to those categories while doing so (Baker Reference Baker and Silverman1997). This is how social order is created and maintained. Ontologically and epistemologically, MCA has a strong affinity with social constructivism and discourse analysis (Bowker and Star Reference Bowker and Star1999), and it is centrally based on the idea that categories are produced in interaction as a result of negotiation. The categories do not exist in any stable form, but their boundaries require the consideration and evaluation of the qualities attached to them. MCA has been used, for example, to analyse how people create identities in talk (Antaki and Widdicombe Reference Antaki and Widdicombe1998; Cummins Reference Cummins1996; Ethier and Deaux Reference Ethier and Deaux1994), how representations of gender are established (Speer and Stokoe Reference Speer and Stokoe2011), and how age categories and identities are negotiated (Kontos Reference Kontos1998; Nikander Reference Nikander2000, Reference Nikander2002; Pietilä and Ojala Reference Pietilä and Ojala2011).
In this study, MCA has been utilised to analyse how older people categorise themselves and their autonomy in relation to their personally perceived and commonly expected age category, functional ability and self-mastery. Empirically, we focused on sections in which interviewees talked about their own or their age peers' ability, inability and desire to do things, as well as about things that either support or restrict those activities. We also took into account Gilleard and Higgs' (Reference Gilleard and Higgs2011b , Reference Gilleard and Higgs2013) notion of ‘otherness' as reflected through its representation within ‘third-person narratives', which importantly creates the boundaries of the group of ‘us' in distinction from ‘others'. The focus of the analysis is thus on the interviewees' negotiations of the boundaries of autonomy, their categorisations of autonomy and autonomous people, and the qualities they attach to these categories.
Negotiating autonomy at the age of 90
The perceived level of autonomy was contemplated and negotiated in relation to three domains: functional ability, independence in decision making and contesting norms of age-appropriateness. All interviewees stressed the value of functional ability, which was considered in terms of being physically capable of managing daily tasks without other people's immediate assistance. This conceptualisation of autonomy comes close to how autonomy has been typically approached in gerontological research (Berg et al. Reference Berg, Hassing, McClearn and Johansson2006; Perrig-Chiello et al. Reference Perrig-Chiello, Perrig, Uebelbacher and Stähelin2006; Schwanen and Ziegler Reference Schwanen and Ziegler2011). Our interviewees based their independence in decision making on material and financial self-sufficiency and the retention of cognitive ability. Autonomy in relation to norms of age-appropriateness was constructed by turning age categories upside down and taking liberties by appealing to old age. Our interviewees stressed these domains differently; while a person might be dependent on others' help in carrying out daily tasks, the person might simultaneously feel autonomous in other respects. It is worth noting that in most cases the interviewees' talk about autonomy was intertwined with talk about their life satisfaction. This highlights the major role that autonomy plays as the basis of life satisfaction among community-dwelling older people.
Functional ability
Health and functional ability were often spoken of as synonyms. In the excerpt below, a 90-year-old woman is being asked what she thinks about good old age.
Yes I am just satisfied. And I'm satisfied that I've been even this healthy. I was able to take care of my husband so that he didn't have to stay in institutional care. And I've been able to take care of myself so far, well you never know. And I'm able to do the housework here, and to cook for myself, and I go out there. (Excerpt 1, female, living alone)
The excerpt shows that standards of autonomy are adapted to the person's age (e.g. being ‘even this healthy’ and ‘so far’ referring to her perceived age), and that they include a comparison with expected levels of health at that age (cf. Jylhä Reference Jylhä1994). Although the female interviewee starts her account with a reference to health, she talks de facto about functional ability. Her satisfaction with her health converges with being able to take care of various everyday tasks without other people's assistance. The tasks mentioned relate to domains of domestic work and care that may be interpreted as gendered ideals of ability. The excerpt implies that not all women of the interviewee's age are able to manage these tasks independently, and autonomy thus rests on the idea of not having fallen into the category of age peers who are no longer capable of living their lives independently.
The majority of interviewees described various practices of selection in relation to daily tasks, most often in relation to mobility. It is worth noting, however, that some interviewees were still in quite good physical condition, and that some of them were notably proud of this. The men in the sample had fewer diseases and functional limitations than their female counterparts, which is in line with many epidemiological studies on the health and functional ability of older people (Berg et al. Reference Berg, Hassing, McClearn and Johansson2006). This also means that men's operational environments were often described as wider, and accordingly their negotiations of autonomy were on a somewhat different level.
I go to the gym three times a week. And then I go dancing. You have to have activities … And I've still got a car, you know, I certainly still drive it, because I still have five years [on my driving licence]. It's in the garage, it's a warm garage. So I can go whenever I want … You know what, I'll go to the dance. (Interviewer: You mean, where's the dance?) It's in [place name]. It starts there at 1.30 pm, a tough boy. (Excerpt 2, male, living alone)
The excerpt can be read as a celebration of the public discourses of active and successful ageing. In his answer to the interviewer's question about his hobbies, the interviewee emphasises his active lifestyle and concludes that one has to be active to maintain a good condition. The interviewee also tells the interviewer that it is easy for him to keep in touch with friends, since he still has a driver's licence and a car. Previous research has shown that driving is associated with activities outside the home as well as satisfaction in experienced mobility options (Siren and Haustein Reference Siren and Haustein2014). In Finland, people aged over 70 have to get a doctor's certificate confirming their ability to drive a car every five years, unless there are medical reasons to seek this confirmation more often. The interviewee is thus able to state proudly that he has certification for another five years, which can be taken as impressive evidence of his exceptionally good physical condition. Simultaneously, this statement can also be interpreted as a manifestation of a preserved masculine identity. Previous research has shown that losing their driving licence is a crisis for many men, as it represents a certain loss of manhood (Davidson Reference Davidson2008).
In addition to the demonstration of general activity levels, the regular practice of going out dancing might also be seen as a gendered activity highlighting one's sex. Going to a dance is a public act, and the speaker gives the impression he is the only man of his age at the dance, which underlines his exceptionality. In both these senses, going out to a dance by driving there in his own car allows the interviewee to conclude that he is ‘a tough boy’. Various dimensions of functional ability are presented here as a means to secure one's autonomy.
An essential element in assessing one's own functional ability based autonomy at the age of 90+ is the notion that things could be worse. Our interviewees repeatedly made comparisons with other people of their age or younger whose health was worse, and who therefore had lost their autonomy.
If you just could pass away in your own bed, so that you were not taken from one place to another. I've been there enough in these care, in hospitals and all them. There are these, so elderly that are dragged from place A to place B. (Excerpt 3, female, living alone)
In the excerpt, the interviewee's current life – and her autonomy in it – is constructed in relation to age peers who have lost their independence due to health problems. The categories of people referred to are those who are physically able and the disabled others. The account simultaneously involves consideration of one's own future. The interviewee says that she would wish to die peacefully in her own bed. Many interviewees expressed a similar wish for a sudden death, before any final decline in their ability to function. Two people in our data talked about voluntary euthanasia, and one mentioned suicide as a solution. This highlights that death seemed not to be one of the things our interviewees fear; what they were afraid of was losing their autonomy, especially their bodily self-control. For our interviewee, the loss of autonomy is evidently associated with being in hospital, lying in bed and being dependent on nursing staff that ‘drag’ older people from one place to another. In many interviews, institutional care was automatically seen as a restriction on personal autonomy, and therefore thinking about hospitalisation caused fear. In a sense, dying in one's own bed instead of in an institution might be taken to represent an ultimate form of autonomy.
Alongside general functional decline, dementia was considered a particularly major threat to autonomy. The man in the next excerpt makes a clear distinction between demented and cognitively able people, and considers his own present life, autonomy and life satisfaction in relation to the category of those with memory disorders:
But for instance those with memory disorders, that's quite a miserable bunch of people. Often they are still physically in good condition. And then relatives come to see them and they don't recognise them any more. Those are such [issues] that make a difference [concerning quality of life]. (Excerpt 4, male, living in senior housing)
The speaker views the demented as ‘quite a miserable bunch of people’; a person can still be in good physical shape while not recognising even his or her own relatives. While the previous interviewee feared losing her physical ability to function and ending up bedridden, this man clearly fears losing his cognitive capacity in spite of a good physical condition. Both situations represent a decline of self-care (Gilleard and Higgs Reference Gilleard and Higgs2000), but the loss of cognitive control is portrayed as more ominous than the loss of physical ability. This is because dementia is in stark conflict with the ideals of autonomy, rationality, emotional mastery and self-control emphasised in the contemporary idea of a man. As Williams, Higgs and Katz (Reference Williams, Higgs and Katz2012) put it, memory disorders represent the collapse of the mental competencies that mark out the reflexive self in the modern world.
Independence in decision making
Independence in decision making was a broad theme in the interviews, ranging from material welfare to freedom from daily obligations. Both men and women considered material resources as a significant factor concerning their perceived autonomy, which is in line with previous research (Plath Reference Plath2008; Schafer, Mustillo and Ferraro Reference Schafer, Mustillo and Ferraro2013). Several interviewees made a comparison between their past and present lives, and concluded that the latter was better in terms of housing, personal finances and living standards among old people in general. It also seemed that men more often considered their new freedom in terms of freedom from their previous economic and work-related responsibilities, whereas women underlined the importance of independent time management. In the excerpt below, a female interviewee has been asked what she considers good old age, and what is required to achieve it.
Health, of course. And then that you have such a living that you're able … I mean that you have a good enough pension that you can pay for all these treatments and all the rest. You go pretty much to your GP [general practitioner in the national health-care service], but it's difficult to get there too. You certainly have to use a private doctor many times, at least I've had to, 'cause you don't go to your GP just like that. (Excerpt 5, female, living in senior housing)
In this excerpt, the significance of health as a domain of autonomy emerges clearly, just as in almost all of the interviews. The interviewee also links health and financial security together. Having enough money (a good pension) makes it possible to take care of one's health. Several interviewees shared this view that public health care is good but not enough (‘you don't go to your GP just like that’). Therefore, having a secure economic position gave more options in decision making. This conclusion is again made based on an implicit division between people who have this opportunity and others who do not.
As noted above, in men's accounts in particular the idea that one's current economic situation was better than in one's past life was based on a consideration of economic pressures during earlier life stages.
Well yes, if you think about middle age and your best age, and about what people have in it. When you're close or over 20 and closer to 35, when children are small, like there's certainly more responsibility then, because you have to arrange housing. And then when you have debt, you have to pay it off. So there's a lot of responsibility. And then again to top it all off there are illnesses, if you just could live a healthy life. And when you have got rid of those other worries, when you have no worries about housing and children, so that's a relief then too. (Excerpt 6, male, living alone)
The interviewee reflects on his current autonomy by distinguishing it from the hectic years of having children and a mortgage. His current phase of life appears as a relief from earlier responsibilities and concerns, including economic worries. This kind of categorisation, between people who have to take care of others and those who do not, was made in many interviews. Men typically approached this from the point of view of finances and their previous breadwinner role in their family. On the other hand, women seemed to contrast their present with their earlier lives from a different angle, emphasising their release from various everyday obligations. The female interviewee in the excerpt below tells the interviewer about the start of her normal day.
First I tend not to get out of bed, because I want to read; at least one and a half hours I read my newspaper there. Well certainly I get up at around nine o'clock and make coffee and such like. But I suppose you somehow like it, 'cause before you had to get up in the morning. It's wonderful when you have no rush in the mornings. Of course you may have plans, but nothing extraordinary. But you don't need to run in the early hours. (Excerpt 7, female, living alone)
In Finland, it is characteristic of the generation of people now in their nineties that throughout their lives women did most of the housework, regardless of whether they were also engaged in paid labour outside the home. This female interviewee's story of her morning activities reflects the numerous domestic obligations of her earlier life. This represents an idea of current autonomy being based on independent time management. The excerpts above thus exemplify an important gender difference in relation to household tasks: while men appreciated freedom from economic obligations, women emphasised freedom from concrete tasks and daily routines. For our interviewee, it is luxurious to read the morning paper for one and a half hours before getting up.
Some men also expressed more abstract values in relation to autonomy and life satisfaction. One of the male interviewees refers to increasing life experience as a significant issue in his feeling of autonomy.
You have to live in accordance with what you consider right. It has to come from your own head. Maybe it helps if you, whatever may come, don't take it all that seriously. And that hasn't happened to me. Of course there have sometimes been various kinds of mess-ups but life teaches you. (Excerpt 8, male, living with spouse)
For this man, autonomy is based on acting according to one's own values. Living like this may sometimes lead to conflicts with other people, but that is just life. Feeling autonomous arises from a kind of relaxed masculinity. One follows one's own path, not avoiding conflicts but not getting too embroiled in them: ‘life teaches you’, like the man said. Another man from our sample provides a less ‘relaxed’ version of the same idea:
When nobody bosses you around, so that you can be [as you want], that makes me feel good. That's how it has been with us [referring to his marriage], and if something has happened and I've had a bad day. It's been sometimes, very seldom, but like if I have fallen down and hit my head. And someone says ‘now go to a doctor’. I've been allowed to decide myself whether I will die right here or must I really go to a doctor. I don't have to go, and I won't, then. It's me who decides what I like and what I do. No one tells me what I can do. Except the wife, who's always bossing. ‘Put socks on when you go out.’ ‘Put a coat on, and a shirt and trousers', I'm not allowed to go without trousers. Underpants I certainly have, but even they should be long johns, and woollen long johns and all the rest. I'm like a warm-blooded stallion and that one there is a cold-blooded mare. (Excerpt 9, male, living with spouse)
This account can be read as an ultra-masculine definition of autonomy: life is good when nobody bosses you around. This message is loud and clear, and the storyline is spiced up with extreme examples of a man being free even to die when he wants, emphasising that autonomy here means an absolute freedom of choice. In the latter part of the account, his wife (who is present during the interview, commenting on the conversation) is humorously portrayed as the only ‘outsider’ who is allowed to enter the man's space.
Our interviewees' strong emphasis of autonomy is in line with previous studies evincing that Finnish old people wish to live in their own homes and want to manage on their own as long as possible (Jolanki Reference Jolanki2009). Similarly, for 30 years or so Finnish social policies have emphasised independent living at home as a priority, regardless of older persons' diminishing functional abilities. A new law called ‘Act on Supporting the Functional Capacity of the Older Population and on Social and Health Services for Older Persons' came into effect on 1 July 2013 (Finlex 2012). The Act clearly expresses that in the future people should age in place in Finland. This means that institutional care will be replaced with home care and sheltered housing.
Contesting norms of age-appropriateness
In addition to freedom from responsibilities and obligations, negotiated autonomy was also tied up with more abstract notions of contesting norms of age-appropriateness. Some interviewees expressed the idea of breaking social norms and expectations.
I have got such a pain-relief plaster there on my back. It was just extended and so I'm a bit … One doctor said that there's dope in it. But I said that couldn't this kind of old granny already be a bit drugged. And she laughed at that. She thought that this old woman has indeed got high. (Excerpt 10, female, living alone)
This woman is playing with the idea that old age provides her with permission to ‘get high’. This is grounded in the notion of being released from daily responsibilities, but also in the social expectations that restrict younger people. This woman creates a distinction between those who have to follow strict social roles and those who are socially free, at least in a relative sense. Our 90–91-year-old interviewees also played with the idea of mixing age categories.
You've got a phone directory so you can check what Raimo's number is. And you can call him and say ‘Raimo, come over, we'll wet our whistles’ cause it's my birthday’. And then we've agreed no gifts. We all have what we need, that's old-fashioned. If you want to make the gang laugh then you buy a lollipop, and bring that. You buy a lollipop for a 90-year-old man, that's a good laugh. You give your gift and give a speech and so forth. (Excerpt 11, male, living alone)
This excerpt speaks of 90-year-old men who already have everything they materially need. Birthday presents therefore become an opportunity to fool around with age categories. It is a good laugh to give a lollipop as a present to a fellow 90-year-old man, accompanied by a formal speech. Giving a lollipop to an age peer has two functions here. Firstly, it might be a kind of protest against celebrating birthdays during old age. A lollipop is an anti-present. Secondly, a lollipop is something usually given to children. The interviewee and his friends understand the constructed nature of age categories and have fun by turning them upside down. It has been suggested (e.g. Salari Reference Salari2006; Salari and Rich Reference Salari and Rich2001) that one of the ways to marginalise older people is to infantilise them. This man demonstrates that older people have their own ways of resisting any attempts to be treated like children. This may also be read as an expression of resistance towards ageist images of old men who take birthday celebrations too seriously, and a representation of a young-at-heart attitude to life (Hurd Reference Hurd1999).
The examples presented above highlight that our interviewees were aware of the existence of age-related norms, and that the features socially attached to the fourth age are indeed a part of their social imaginary (Gilleard and Higgs Reference Gilleard and Higgs2013). It is well known that representations of old age affect the individual self-image of older persons (Coupland Reference Coupland2009; Häggblom-Kronlöf et al. Reference Häggblom-Kronlöf, Hultberg, Eriksson and Sonn2007; Queniart and Charpentier Reference Queniart and Charpentier2011; Stephenson et al. Reference Stephenson, Wolfe, Coughlan and Koehn2000). Our interviewees seemed to recognise the normative nature of age categories, and consequently they intentionally broke them. Very old age was used an excuse to do socially questionable things; but on the other hand, by turning age categories around, our respondents actually confirmed the existence of these age-related norms. Breaking age-related norms, even imaginatively, might be emotionally important, as it creates feelings of autonomy. For example, interpreting the use of medication in terms of ‘getting high’ offers an opportunity to feel autonomous and take joy in life, even when one is forced to use medication due to illness. The redefinition of the medication – from something that has been prescribed to something that is actively used for fun – allows the speaker to denote agency and autonomy in a situation which otherwise might be seen as restricting them.
Discussion
It is clear that for most people physical independence decreases due to functional limitations at the age of 90+ years. Nonetheless, it has been noted that older people are typically reluctant to describe themselves as frail and to adopt that identity (Degnen Reference Degnen2007; Gilleard and Higgs Reference Gilleard and Higgs2011b ; Heikkinen Reference Heikkinen2004; Hurd Reference Hurd1999). In Kaufman's (Reference Kaufman1986) study, older people psychologically avoided feelings of becoming frail by dealing with specific problems and disabilities as they occurred, just as they had been doing throughout their lives. In line with this notion, it was only in rare cases that our interviewees talked about themselves as frail. On the other hand, however, they expressed realistic views of their diminishing capacities for independent living and the functional limitations they perceived in their everyday lives. This highlights an important difference between the two concepts functional decline and frailty. At the age of 90, people can hardly ignore their own physical ageing, particularly when being interviewed by a health scientist. Therefore, by talking about their functional limitations instead of frailty they maintain a competent and autonomous identity while simultaneously avoiding falling into the trap of denying their ageing (Kaufman Reference Kaufman1986; Tulle Reference Tulle2007).
In this study, we found that negotiations of autonomy were intertwined with talk about life satisfaction, which confirms the finding of previous studies that the perceived level of autonomy plays a constitutive role in perceptions of life satisfaction among the oldest old (Enkvist, Ekström and Elmståhl Reference Enkvist, Ekström and Elmståhl2012; King et al. Reference King, Yourman, Ahalt, Eng, Knight, Pérez-Stable and Smith2012). However, our study also shows that in comparison with other life stages, for 90- and 91-year-olds autonomy is much more a matter of reflexive consideration and negotiation. At that age many issues that were previously more or less self-evident aspects of autonomy, such as independent mobility and the handling of daily tasks at home, are no longer so self-evident. Therefore, the issues on which one's feeling of autonomy is based become matters for reconsideration which is in line with previous studies concerning coping in later life (Ben-Zur Reference Ben-Zur2002; Lloyd et al. Reference Lloyd, Calnan, Cameron, Seymour and Smith2014; Ouwehand, de Ridder and Bensing Reference Ouwehand, de Ridder and Bensing2007; Sarvimäki and Stenbock-Hult Reference Sarvimäki and Stenbock-Hult2000). Being able to cook, go grocery shopping or have a driving licence become important signs of autonomy. In some cases, these are not necessarily even tied up with the activities in which the people are actually engaged. Owning a driving licence and a car might be important even if an older person does not actually drive, as it represents a potential to act autonomously.
All this suggests that aspects of autonomy become more subtle. At the same time, feeling autonomous does not necessitate a long list of independent activities in various life contexts, but rather even single and potential factors may be enough. A feeling of autonomy may be based on one's ability to make choices and decisions autonomously, even if one is dependent on other people for physical tasks. In a study by Lloyd et al. (2014: 13), some older persons did paradoxically base their feeling of autonomy on the help of family members and family members considered their help as ‘helping him/her to keep a bit of independence’. Our study confirms the redefinition of activities and abilities that are considered the prerequisites of an autonomous person. The redefinition may also concern expectations within the social environment. Questioning and breaking age-related norms serves to manifest full membership of social interaction as well as an ability to consider reflexively and even manipulate the social order.
Several researchers have claimed that in the gerontological literature gender is among the less-highlighted perspectives on the experienced lives of older people (e.g. Calasanti Reference Calasanti2004; Twigg Reference Twigg2004). In quantitative analyses, gender is typically one of the most common background variables. Despite this, older people are often approached just as ‘old’, i.e. in terms of their age, but not explicitly as older women and men. This may reflect a wider tendency to see older people as somewhat genderless (Fleming Reference Fleming1999), which may be even more heavily emphasised for the oldest olds. In Thompson's (Reference Thompson2006: 633) words, ‘aging overshadows gender’. Our study reveals that gendered ideals and values play an important role in our interviewees' conceptions of autonomy. Issues that are considered to produce autonomy centrally revolve around the gendered roles and life domains of women and men. To be independently able to take care of her household, even partly, creates feelings of autonomy for a 90-year-old woman as a woman. For a man of the same age, having a car and a driving licence may produce similar feelings of control over his own life. Gender thus matters in the lives of nonagenarians. It seems that the key point of autonomy-based life satisfaction is to make choices in order to continue the habitual aspects of gendered life. In order to support the life satisfaction of the oldest old, it might be important to consider how to promote gendered aspects of life in old age in order to support the continuity of their gendered identities.
Chris Gilleard and Paul Higgs (Reference Gilleard and Higgs2013) have claimed that in contemporary societies later life might have lost much of its previous coherence. Instead, diversity, difference and inequality have become more salient. This also has an effect on how the oldest people understand their own position in age and life-stage categories and hierarchies. Our analysis showed that our interviewees were clearly aware of the diversity within their own age group, particularly in terms of mental and physical functioning. However, it also showed that these diversities were actively mobilised as resources for positive self-identification, which in turn indirectly produced hierarchies, marginalisations and even inequalities within the group of oldest old.
From the perspective of how people ‘do’, ‘perform’ or ‘accomplish’ their age (Calasanti Reference Calasanti, Biggs, Lowenstein and Hendricks2003; Laz Reference Laz2003), our study highlighted features similar to those found in studies exploring the ways in which middle-aged and even younger people identify themselves in terms of age categorisations (e.g. Nikander Reference Nikander2002, Reference Nikander2009; Pietilä and Ojala Reference Pietilä and Ojala2011; Pietilä et al. Reference Pietilä, Ojala, King and Calasanti2013). When considering their own ageing, people usually make comparisons and contrasts between their past and current selves and both younger and older generations (Nikander Reference Nikander2009). This age-based categorisation creates groups that we might call age-others. For people in their nineties this is not equally possible, as there exists no clear older age group ahead of them. Therefore, our respondents' comparisons were not based on age, as in younger age groups, but on the distinctions they made in terms of ability to function. Frail age peers replaced the ‘older’, as a point of comparison. This functional distinction was constructed between people who can take care of themselves and ‘others' who have lost their autonomy because of ageing. Age-others became ability-others. This highlighted that what actually matters was perceived autonomy, which did not apply only to functional ability, but also to other forms of self-management. Regarding autonomous decision making, the ‘others' were those of the same age (or even younger) who had lost their autonomy due to placement in a care institution or having memory disorders.
Regarding both their chronological age and their functional status, people aged over 90 years can hardly be considered to represent the third age. Our analysis showed that, on the one hand, they could easily be seen as representatives of the fourth age with the features of dependence and frailty socially attached to this category. On the other hand, membership of the group of the fourth agers was a much more nuanced issue. This was because our interviewees themselves were aware both of their own membership of the oldest age group and of the diversity that characterises that group. While the group is generally characterised by functional decline and increasing dependence, there are substantial individual differences in how people cope with age-related changes in their everyday lives (Ben-Zur Reference Ben-Zur2002; Ouwehand, de Ridder and Bensing Reference Ouwehand, de Ridder and Bensing2007; Sarvimäki and Stenbock-Hult Reference Sarvimäki and Stenbock-Hult2000). It could thus be said that currently the existence of diversity related to the fourth age is an essential feature of the social imaginary of age and ageing among the oldest old. The awareness of diversity within age groups makes it possible to create subcategories and distinctions within the group of the oldest old. The making of these distinctions – between the dependent but autonomous, on the one hand, and the dependent and no longer autonomous, on the other – results in the construction of a new location which has characteristics of both the third and fourth age.
Our interviewees considered themselves as chronologically old, but they did not regard themselves as fitting ‘real old age’ due to their perceived level of autonomy. Our analysis is in line with Lloyd et al. (Reference Lloyd, Calnan, Cameron, Seymour and Smith2014) who found out that ‘perseverance’ was the attitude of older people struggling somewhere between the third and fourth ages. The fourth age was postponed by continuous reflexive processes dealing with current and earlier lifecourse experiences and changes in individual health and abilities. Perseverance was subtly connected with adaptation since people did recognise the upcoming loss of agency. They used phrases like ‘as long as' and ‘so far’ a lot, revealing that their perseverance was realistic.
Our analysis shows that, in line with Gilleard and Higgs' (Reference Gilleard and Higgs2010, Reference Gilleard and Higgs2011a , Reference Gilleard and Higgs2013) ideas, the fourth age indeed plays an essential role in our social imaginary of ageing. However, it also shows that while the concepts of the third and fourth age are illuminating on a societal level, their analytic applicability is limited in the empirical analysis of how people consider their own positions within age and life-stage categories. Together with Lloyd et al. (Reference Lloyd, Calnan, Cameron, Seymour and Smith2014), our study evinces that the fourth age, as we usually understand it in the sense of decline and loss of agency, is hard to define in terms of individual experience. Autonomy is always a negotiable characteristic. If a person – regardless of his or her chronological age or functional abilities – defines themself as ‘autonomously dependent’ or ‘dependently autonomous', they do not belong to the group of the fourth agers. The diversity within the group of the oldest old will presumably widen as that age group grows rapidly in many countries. This increasing diversity creates blind spots and grey areas within age categories that the concepts of the third and fourth age seem unable to recognise.
Acknowledgements
The authors would like to thank the University of Tampere for providing facilities to carry out this study. We also want to thank people conducting the Vitality 90+ study in the School of Health Sciences and Gerontology Research Center at the University of Tampere for the opportunity to use their data. A statement of ethical approval was not required for this study.
